Hypogonadotropic hypogonadism (HH) occurs in approximately one-third of obese and type 2 diabetic men. Considering that there are 24 million diabetic and 100 million obese people, of whom half are males, obesity and type 2 diabetes potentially constitute the major cause of hypogonadism in the population. We hypothesize that 1) HH in obese and type 2 diabetic men is associated with decreased insulin sensitivity, increased fat tissue mass, decreased lean body mass, increased inflammatory and oxidative stress, impaired sexual function and depressed mood as compared to diabetic and obese men with normal testosterone concentrations; and that 2) testosterone replacement for 24 weeks in men with HH leads to an improvement in these parameters. Our proposed study would be the first prospective, randomized trial to comprehensively evaluate the effect of HH on insulin sensitivity, body composition, inflammatory and oxidative indices in obese and type 2 diabetic subjects and the effect of six months of T replacement on these parameters. The study will have 2 arms (obese and type 2 diabetic arm) with 120 subjects in each arm. Each arm will have 60 men with HH and 60 men with normal testosterone concentrations (eugonadal men). Insulin sensitivity will be assessed by hyperinsulinemic-euglycemic clamps. Subcutaneous fat mass and lean body mass will be measured by DEXA and intra-abdominal (visceral) fat mass by MRI. All subjects will undergo hyperinsulinemic-euglycemic clamp, MRI, DEXA and give blood and urine samples (for measurement of inflammatory and oxidative stress) at baseline. Men with HH will then be randomized to receive testosterone or placebo gel for a total of 24 weeks. These men will undergo hyperinsulinemic- euglycemic clamps and give blood and urine samples for inflammation and oxidative stress at 4 weeks and 24 weeks. MRI and DEXA examinations will be carried out at 24 weeks again in men with HH. The primary endpoint of the study is to define a difference in whole body glucose uptake during hyperinsulinemic- euglycemic clamps between hypogonadal and eugonadal patients at baseline and an increase in glucose uptake in HH subjects after treatment with testosterone for 24 weeks. 30 subjects per group (testosterone and placebo gel each) will provide adequate power (0.8) to detect a significant difference of 10% in whole body glucose uptake. Therefore there will be 60 men with HH in each arm. For baseline comparisons, 60 men with normal testosterone concentrations will also be needed in each arm. Thus there will be 120 men in each arm and a total of 240 subjects in the study. PUBLIC HEALTH RELEVANCE: The decrease in insulin resistance after treatment with testosterone has important implications for both glucose homeostasis in type 2 diabetic patients and for prevention of type 2 diabetes in obese men. The decrease in inflammation with testosterone treatment may have relevance for atherosclerosis and cardiovascular risk.